Item Based Collaborative Filtering Recommendation Algorithms

1
Item Based Collaborative Filtering Recommendation
Algorithms

• Badrul Sarwar,
• George Karpis,
• Joseph KonStan,
• John Riedl
• (UMN)

Presenter Yu-Song Syu
p.s. slides adapted from http//www.cs.umd.edu
/samir/498/CMSC498K_Hyoungtae_Cho.ppt
2
Introduction

• Recommender Systems Apply knowledge discovery
  techniques to the problem of making personalized
  recommendations for information, products or
  services, usually during a live interaction
• Collaborative Filtering Builds a database of
  users preference for items. Thus, the
  recommendation can be made based on the neighbors
  who have similar tastes

3
Collaborative Filtering in our life
4
Collaborative Filtering in our life
5
Collaborative Filtering in our life
6
Motivation of Collaborative Filtering (CF)

• Need to develop multiple products that meet the
  multiple needs of multiple consumers
• Recommender systems used by E-commerce
• Multimedia recommendation
• Personal tastes matters

Key
7
Basic Strategies

• Predict and Recommend
• Predict the opinion how likely that the user
  will have on the this item
• Recommend the best items based on
• the users previous likings, and
• the opinions of like-minded users whose ratings
  are similar

8
Traditional Collaborative Filtering

• Nearest-Neighbor CF algorithm (KNN)
• Cosine distance
• For N-dimensional vector of items, measure two
  customers A and B

9
Traditional Collaborative Filtering

• If we have M customers, the complexity will be
  O(MN)
• Reduce M by randomly sampling the customers
• Reduce N by discarding very popular or unpopular
  items
• Can be O(MN), but

10
Clustering Techniques

• Work by identifying groups of consumers who
  appear to have similar preferences
• Performance can be good with smaller size of
  group
• May hurt accuracy while dividing the population
  into clusters

But
11
How about a Content based Method?

• Given the users purchased and rated items,
  constructs a search query to find other popular
  items
• For example, same author, artist, director, or
  similar keywords/subjects
• Impractical to base a query on all the items

But
12
User-Based Collaborative Filtering

• Algorithms we looked into so far
• 2 challenges
• Scalability Complexity grows linearly with the
  number of customers and items
• Sparsity The sparsity of recommendations on the
  data set
• Even active customers may have purchased well
  under 1 of the total products

13
New Approaches?
14
Item-to-Item Collaborative Filtering

• No more matching the user to similar customers
• build a similar-items table by finding that
  customers tend to purchase together
• Amazon.com used this method
• Scales independently of the catalog size or the
  total number of customers
• Acceptable performance by creating the expensive
  similar-item table offline

15
Item-to-Item CF Algorithm

• O(N2M) as worst case, O(NM) in practical

16
Item-to-Item CF AlgorithmSimilarity Calculation
Computed by looking into co-rated items only.
These co-rated pairs are obtained from different
users.
17
Item-to-Item CF AlgorithmSimilarity Calculation

• For similarity between two items i and j,

18
Item-to-Item CF AlgorithmPrediction Computation

• Recommend items with high-ranking based on
  similarity

19
Item-to-Item CF AlgorithmPrediction Computation

• Weighted Sum to capture how the active user rates
  the similar items
• Regression to avoid misleading in the sense that
  two rating vectors may be distant yet may have
  very high similarities

20

• The item-item scheme provides better quality of
  predictions than the user-user scheme
• Higher training/test ratio improves the quality,
  but not very large
• The item neighborhood is fairly static, which can
  be pre-computed
• Improve the online performance

21
Conclusion

• Presented and evaluated a new algorithm for
  CF-based recommender systems
• The item-based algorithms scale to large data
  sets and produce high-quality recommendations

22
Item-to-Item CF AlgorithmPrediction Computation

• Weighted Sum to capture how the active user rates
  the similar items
• Regression to avoid misleading in the sense that
  two similarities may be distant yet may have very
  high similarities

23
References

• E-Commerce Recommendation Applications
  http//citeseer.ist.psu.edu/cache/papers/cs/14532/
  httpzSzzSzwww.cs.umn.eduzSzResearchzSzGroupLenszS
  zECRA.pdf/schafer01ecommerce.pdf
• Amazon.com Recommendations Item-to-Item
  Collaborative Filtering http//www.win.tue.nl/lar
  oyo/2L340/resources/Amazon-Recommendations.pdf
• Item-based Collaborative Filtering Recommendation
  Algorithms
• http//www.grouplens.org/papers/pdf/www10_sarwar.
  pdf